youngstructural
Structural
- Aug 17, 2004
- 713
I'm going to be entirely upfront here and point out that I am a relatively new Structural Intern. As such I often find myself surprised by accepted practice and the occational rediculous "generally accepted practice" that doesn't really work at all.
My question is this: How do you handle the refusal of the wood truss manufacturers to design and take responsibility for the permanent truss bracing system?
A little bit on my overall understanding, perhaps for edification, perhaps for rebuttal, but here it is:
Wood trusses are typically spec'd by the building designer as trusses @ x dist. o.c. and in a given direction, with requisite design loads. From there the truss supplier runs their truss plate manufacturer produced software following (if you're lucky) your truss layout and inputing your required loads.
The software provided by the plate manufacturer outputs locations of lateral support for given members. This is of course to reduce the effective unsupported length of members in compression and enable a more efficient design. Also, they are looking to eliminate the need for t-bracing of members, since this makes staking the trusses more complex and difficult, if not (as is sometimes the case) impossible.
From here, the contractors (nearly always) interpret the bracing requirement to mean simply installing the bracing shown on the drawings (typically 2x3 or 2x4 perpendicular to the truss between similar trusses throughout) between the trusses and leaving it there. I would question the ability to get just a small number of nails to work, let alone take the loads which would normally be present. And let's not even get into a discussion of load paths and/or the fact that you can't connect two laterally unstable systems together with a single simple connection and magically call them braced. Two laterally unsupported wall are still laterally unsupported if you connect them together with a simple 2x4 and a few nails, and so are trusses!
Nearly all of the literature out there seems to be very biased in favour of the truss plate manufacturers who seem to just want this problem to go away without their further consideration, expense or effort. A really good example of this can be found in Dr. Frank E. Woeste's paper "Wood Truss Bracing" which can be found at .
obviously a very competent man, I find no fault in his presentation of the problem at hand, but I do find a great deal of fault with the paper. How is the engineer of record on the project supposed to design the lateral bracing system without knowing the types of trusses being used? And if this is the format we are to follow, who will pay for the additional consulting fees? Why should we recommend or specify the use of wood trusses to a client when the steel truss manufacturers have been handling this problem and giving guidance on it for decades? And really, it is difficult enough to keep the truss manufacturers from deviating from our truss layouts (which can of course mean a shift in point loads and all sorts of unforseen complications).
I have spent some time looking into this design issue, since I have no wish to specify a system without understanding it's functioning, and have come to two possible reasons for why this does indeed seem to work in the real world:
1. The wood sheathing screwed to the trusses is acting as shear panels, which are taking loads from the various members being braced together as simple beams. The loads are thus transfered down to the sidewalls and through the walls to the foundation.
2. The drywall on the underside of the trusses is effecting the same load path as above, simply on the horizontal.
Now, there are many things in our everyday lives that do simply work and have worked and will probably continue to wok for a long time. Hoewever, that is not what engineers do or what our profession is about. We are here to ensure the safety of a structure. To give the client the most economical design possible. To defend the safety of the layman against unsafe or unscrupulous workmanship. And even if this system does work, the "shear panel" and the "underside drywall" suggestions are only qualitative models of what makes this work, not quantitative solutions for the given system. And I certainly have no interest in being the one left holding the short end of the rope when something does "just happen" to go wrong. On top of it all, without further reasearch or guidance, it would be impossible to predict when the current "it just works" solution could fail.
Hopefully some of you guys can explain all of this to me. And if you've gotten this far, thank you very much for sticking it out. Sorry for the long read.
My question is this: How do you handle the refusal of the wood truss manufacturers to design and take responsibility for the permanent truss bracing system?
A little bit on my overall understanding, perhaps for edification, perhaps for rebuttal, but here it is:
Wood trusses are typically spec'd by the building designer as trusses @ x dist. o.c. and in a given direction, with requisite design loads. From there the truss supplier runs their truss plate manufacturer produced software following (if you're lucky) your truss layout and inputing your required loads.
The software provided by the plate manufacturer outputs locations of lateral support for given members. This is of course to reduce the effective unsupported length of members in compression and enable a more efficient design. Also, they are looking to eliminate the need for t-bracing of members, since this makes staking the trusses more complex and difficult, if not (as is sometimes the case) impossible.
From here, the contractors (nearly always) interpret the bracing requirement to mean simply installing the bracing shown on the drawings (typically 2x3 or 2x4 perpendicular to the truss between similar trusses throughout) between the trusses and leaving it there. I would question the ability to get just a small number of nails to work, let alone take the loads which would normally be present. And let's not even get into a discussion of load paths and/or the fact that you can't connect two laterally unstable systems together with a single simple connection and magically call them braced. Two laterally unsupported wall are still laterally unsupported if you connect them together with a simple 2x4 and a few nails, and so are trusses!
Nearly all of the literature out there seems to be very biased in favour of the truss plate manufacturers who seem to just want this problem to go away without their further consideration, expense or effort. A really good example of this can be found in Dr. Frank E. Woeste's paper "Wood Truss Bracing" which can be found at .
obviously a very competent man, I find no fault in his presentation of the problem at hand, but I do find a great deal of fault with the paper. How is the engineer of record on the project supposed to design the lateral bracing system without knowing the types of trusses being used? And if this is the format we are to follow, who will pay for the additional consulting fees? Why should we recommend or specify the use of wood trusses to a client when the steel truss manufacturers have been handling this problem and giving guidance on it for decades? And really, it is difficult enough to keep the truss manufacturers from deviating from our truss layouts (which can of course mean a shift in point loads and all sorts of unforseen complications).
I have spent some time looking into this design issue, since I have no wish to specify a system without understanding it's functioning, and have come to two possible reasons for why this does indeed seem to work in the real world:
1. The wood sheathing screwed to the trusses is acting as shear panels, which are taking loads from the various members being braced together as simple beams. The loads are thus transfered down to the sidewalls and through the walls to the foundation.
2. The drywall on the underside of the trusses is effecting the same load path as above, simply on the horizontal.
Now, there are many things in our everyday lives that do simply work and have worked and will probably continue to wok for a long time. Hoewever, that is not what engineers do or what our profession is about. We are here to ensure the safety of a structure. To give the client the most economical design possible. To defend the safety of the layman against unsafe or unscrupulous workmanship. And even if this system does work, the "shear panel" and the "underside drywall" suggestions are only qualitative models of what makes this work, not quantitative solutions for the given system. And I certainly have no interest in being the one left holding the short end of the rope when something does "just happen" to go wrong. On top of it all, without further reasearch or guidance, it would be impossible to predict when the current "it just works" solution could fail.
Hopefully some of you guys can explain all of this to me. And if you've gotten this far, thank you very much for sticking it out. Sorry for the long read.
